Fe absorption is affected by many dietary factors. The objective of the present study was to measure the effects of high v. low content of vitamin C, meat and phytic acid in whole diets with Fe-fortified bread on the efficacy of Fe absorption. Thirty-two healthy women with low Fe stores were randomised to three groups, each of which was given two of six test diets containing either low/high amounts of vitamin C, meat or phytic acid, respectively, in a cross-over design. Each diet was served throughout a 5 d period. Fe-fortified rye bread, extrinsically labelled with 59Fe, was given with all main meals. Fe absorption was determined from whole-body counter measurements of 59Fe retention. The fractional non-haem Fe absorption (corrected to a 40% standard absorption by measurements from the reference dose) was 1·9% v. 3·4% (P=0·04) for the low/high vitamin C diets, 3·0% v. 3·5% (P=0·58) on the low/high meat diets and 4·9% v. 3·8% (P=0·24) on the low/high phytic acid diet, respectively. The total Fe absorbed (geometric mean with standard error) varied from 0·43 (se 0·11) mg from the diet with lowest bioavailability to 1·09 (se 0·18) mg from the diet with highest bioavailability (P<0·001). The present whole-diet study indicates that diet composition is a strong predictor of Fe absorption. In the diet with a low content of enhancers and a high content of inhibitors, vitamin C improved non-haem Fe absorption. The total Fe absorption varied 2·5-fold after small alterations of the content of enhancers and inhibitors in the diet.